Re: RMAN backup piece validation

From: Gus Spier <>
Date: Sat, 28 Sep 2013 11:22:00 -0400
Message-ID: <>

Oracle, Solaris 10
We are also clawing our way out of a situation where it appears the SAN introduces corruption into the datafiles, the undo tablespaces, the temp file, the backup pieces, and occasionally, the archive logs.

To be fair, the SAN was "cooked" when an uninformed individual reset the thermostat on the computer floor. Excessive heat seems to have had an unpleasant effect on the disks and controllers. Our hardware/sysadmin teams have been consistently defeated by re-appearance/introduction of corruption after each change of disk.

Of course, I'm home on a Saturday morning instead of hovering over my workplace keyboard, so I cannot accurately reproduce the error messages. Corruption is identified by running the RMAN | recover check logical database command. It is fixed by restoring from tape. If it wasn't an iterative process, it would have been a piece a cake.



On Fri, Sep 27, 2013 at 5:14 PM, Hariharan, Abhishek <>wrote:

> Did you open a SR with MOS? What did they say about it?
> And did you mean that you tried the "restore...validate" after getting the
> ORA-600 errors in the actual restores (same backup pieces) and it went
> smooth?
> Regards
> Abhishek
> -----Original Message-----
> From: []
> On Behalf Of Dragutin Jastrebic
> Sent: Saturday, September 28, 2013 2:17 AM
> To: Jared Still
> Cc:
> Subject: Re: RMAN backup piece validation
> Thank you for your followup , (in fact I was 100% sure that someone will
> ask for more details) As much as I remember it was something ugly like
> ORA-0600 signalled during the real restore phase, which did not happened
> during the restore validate.
> We were able to reproduce the same problem several times.
> Finally we took the previous backup
> Dragutin
> --

Received on Sat Sep 28 2013 - 17:22:00 CEST

Original text of this message